1 New approaches to enhance pollutant removal in artificially 2 aerated wastewater treatment systems 3 4 Andrew I. Freeman a,, Ben W.J. Surridge a*, Mike Matthews b, Mark Stewart c, Philip M. Haygarth a. 5 a Lancaster Environment Centre, Lancaster University, Lancaster, LA1 4YQ, UK. 6 b Peak Associates Environmental Consultants ltd, Lancaster Office, Lancaster Environment Centre, Lancaster University, Lancaster, LA1 7 4YQ, UK. 8 c Manchester Airport Group Plc, Manchester Airport, Water Services Department, Building 30, M90 1AA, UK. 9 * Corresponding Author. Tel: +44 (0) 1524 594516. Email: [email protected] (B.W.J. Surridge). 10 11 Abstract 12 13 Freshwater ecosystems sustain human society through the provision of a range of services. 14 However, the status of these ecosystems is threatened by a multitude of pressures, including point 15 sources of wastewater. Future treatment of wastewater will increasingly require new forms of 16 decentralised infrastructure. The research reported here sought to enhance pollutant removal 17 within a novel wastewater treatment technology, based on un-planted, artificially aerated, 18 horizontal subsurface flow constructed wetlands. The potential for these systems to treat de-icer 19 contaminated runoff from airports, a source of wastewater that is likely to grow in importance 20 alongside the expansion of air travel and under future climate scenarios, was evaluated. A new 21 configuration for the delivery of air to aerated treatment systems was developed and tested, based 22 on a phased-aeration approach. This new aeration approach significantly improved pollutant 23 removal efficiency compared to alternative aeration configurations, achieving > 90 % removal of 24 influent load for COD, BOD5 and TOC. Optimised operating conditions under phased aeration were 25 also determined. Based on a hydraulic retention time of 1.5 d and a pollutant mass loading rate of 1 -1 26 0.10 kg d⁻¹ m⁻² BOD₅, > 95 % BOD5 removal, alongside final effluent BOD5 concentrations < 21 mg L , -1 27 could be achieved from an influent characterised by a BOD5 concentration > 800 mg L . Key controls 28 on oxygen transfer efficiency within the aerated treatment system were also determined, revealing 29 that standard oxygen transfer efficiency was inversely related to aeration rate between 1 L and 3 L 30 min-1 and positively related to bed media depth between 1,500 mm and 3,000 mm. The research 31 reported here highlights the potential for optimisation and subsequent widespread application of 32 the aerated wetland technology, in order to protect and restore freshwater ecosystems and the 33 services that they provide to human society. 34 35 Statement of Contributions: AF conceived the research, led the design and implementation of the experiments, analysed samples, 36 undertook statistical analysis and co-led the writing of the manuscript. BS contributed to the conception of the research and design of the 37 experiments and co-led the writing of the manuscript with AF. PH, MS and MM supported the design and implementation of the research 38 and edited earlier versions of the manuscript. 39 40 Key Words 41 Aerated constructed wetlands 42 De-icer contaminated runoff 43 Phased artificial aeration 44 Oxygen transfer efficiency 45 Organic pollutants 46 Freshwater ecosystems 47 48 49 50 51 2 52 1. Introduction 53 54 Freshwater ecosystems provide services that are critical for human society (Dodds et al., 2013, 55 UNESCO, 2015, Durance et al., 2016). However, these ecosystems also face diverse pressures 56 resulting from population growth, urbanisation, industrial development and a changing global 57 climate (Vorosmarty et al., 2000, Ormerod et al., 2010, Vorosmarty et al., 2010, Matthews, 2016). In 58 consequence, contemporary rates of degradation within freshwater ecosystems significantly exceed 59 historical rates, but also contemporary rates of degradation within other ecosystems (Barnosky et 60 al., 2011, Valiente-Banuet et al., 2015). The changes in ecosystem structure and function that are 61 associated with degradation threaten the integrity of freshwater ecosystems, but also constrain the 62 potential for human society to benefit from the services that could potentially be provided by 63 freshwaters (Gleick, 1998, World Health Organization, 2015). 64 Therefore, there is a growing imperative to protect and restore the status of freshwater 65 ecosystems globally. Chemical water quality is a fundmanetal control on ecosystem status and 66 remains subject to significant anthropogenic pressure (Smith and Schindler, 2009, Schindler, 2012, 67 Malaj et al., 2014, Jekel et al., 2015, Van Meter et al., 2016). Point sources have been recognised as a 68 major contributor of pollutants to freshwaters for several decades in many countries (EEA, 2007, 69 DEFRA, 2012, EEA, 2015), with centralised or decentralised wastewater treatment systems being 70 widely used to improve the quality of wastewater entering freshwaters from point sources. 71 However, the energy demands, greenhouse gas emissions and whole life costs associated with many 72 traditional wastewater treatment technologies are subject to increasing scrutiny (Henriques and 73 Catarino, 2017, Rajasulochana and Preethy, 2016). Alternative treatment technologies, characterised 74 by relatively low energy consumption, by simple operating principles and by minimal whole life costs 75 are increasingly required. Such technologies provide a potentially more sustainable means of 76 protecting or enhancing ecosystem services compared to conventional wastewater treatment 77 technologies. Further, such technologies would support enhanced treatment of point sources in 3 78 countries where significant investment in centralised wastewater treatment infrastructure cannot be 79 made, alongside the treatment of smaller, micro-point sources of wastewater for which traditional 80 technologies may be inappropriate or disproportionately costly. In this context, the research 81 reported here developed a novel treatment technology for wastewater, based on un-planted, 82 artificially aerated, horizontal subsurface flow (HSSF) constructed wetlands. It is recognised that the 83 treatment systems considered in this research do not include the vegetation planting schemes that 84 are common in many constructed wetland designs. This reflects the specific application of the 85 technology to the treatment of wastewater at airports, for which planted systems are potentially 86 inappropriate (see below). However, the HSSF constructed wetland terminology is used within this 87 paper, reflecting the fact that in hydraulic, microbial and bed media geochemical terms, the 88 treatment systems described here share many features that are common with planted HSSF 89 constructed wetlands. 90 Treatment technologies that rely on natural, passive pollutant degradation processes, including 91 constructed wetlands, offer environmental and economic advantages over many traditional 92 wastewater treatment systems (Castro et al., 2005, Vymazal et al., 2006, Kadlec and Wallace, 2009, 93 Vymazal and Kröpfelová, 2009, Freeman et al., 2015, Wu et al., 2015). However, the availability of 94 dissolved oxygen (DO) is frequently a fundmanetal limit on pollutant removal within traditional 95 constructed wetland designs (Wallace et al., 2007, Kadlec and Wallace, 2009, Nivala et al., 2013). In -2 -1 -2 -1 96 saturated HSSF wetlands, 0.12 g m d – 12.11 g m d O2 may be transported into a system 97 through the combination of direct diffusion from the atmosphere and diffusion from the sub-surface 98 root network of wetland vegetation (Armstrong et al., 1990, Brix and Schierup, 1990, Brix, 1997, 99 Bezbaruah and Zhang, 2005, Nivala et al., 2013). Vertical flow constructed wetlands achive higher -2 -1 -2 -1 -2 -1 100 diffusion rates of 28.4 g m d – 156 g m d O2 for saturated systems and up to 482 g m d O2 in 101 fill and drain systems, primarily due to the draw down of air into the bed during sequential filling and 102 draining of wastewater through the wetland substrate (Cooper, 2005, Fan et al., 2013a, Nivala et al., 103 2013). However, the rate of DO supply via these mechanisms within traditional constructed wetland 4 104 designs is often negligible when compared to the rate of DO consumption associated with many raw 105 wastewaters (Nivala et al., 2013). Whilst anaerobic respiration of some pollutants occurs, the 106 resulting pollutant removal rates are often lower than under aerobic conditions, meaning that 107 treatment efficiency is significantly reduced (Huang et al., 2005, Ouellet-Plamondon et al., 2006, Fan 108 et al., 2013b, Nivala et al., 2013, Murphy et al., 2015, Uggetti et al., 2016). The need to improve rates 109 of DO supply in order to enhance pollutant removal in traditional constructed wetlands has led to 110 the development and commercialisation of aerated wetlands for a range of applications across the 111 globe (Wallace, 2001). Aerated wetlands involve the active supply of DO into a self-contained, 112 media-filled treatment bed, to maintain aerobic conditions within the wetland by meeting the DO 113 demand exerted by wastewater. With sufficient DO supplied to the system through aeration, the 114 role of wetland vegetation root transfer for this purpose is significantly reduced and systems can 115 remain un-planted to serve applications in which attacting wildlife is undesirable. However, there is 116 currently no recognised design standard for aerated wetlands (Nivala et al., 2013), alongside limited 117 empirical